摘要
提出一种基于模糊聚类和滑动模控制的模糊过模型控制方法，并将其应用于动力学 方程未知的机械手轨迹控制．首先，采用Ｃ均值聚类算法构造两关节机械手的高木－关野 （Ｔ－Ｓ）模糊模型，并由此构造模糊系统的逆模型．然后，在提出的模糊逆模型控制结构中， 离散时间滑动模控制和时延控制（ＴＤＣ）用于补偿模糊建模误差和外扰动，保证系统的全局 稳定性并改进其动态和稳态性能．系统的稳定性和轨迹误差的收敛性可以通过稳定性定理来 证明．最后，以两关节机械手的轨迹跟随控制为例，揭示了该设计方法的控制性能．
In this paper, a novel fuzzy inverse model control using fuzzy clustering and sliding mode control (SMC) is proposed for the trajectory tracking of robotic manipulators having unknown dynamics. The fuzzy model of a two-link manipulator is built based on the derivation of TS models using fuzzy C-means clustering algorithm and the robot fuzzy inverse model is obtained. Then, in the proposed control framework of fuzzy inverse model, sliding mode control and the time-delay control approach are used to compensate for the fuzzy modeling errors and disturbances, thus the system stability is guaranteed and its tracking performance is improved. The system stability and convergence of tracking errors are proven by stability theory. Finally, an example for the trajectory tracking of a two-link manipulator illustrates the performance improvement of the proposed control approach.
出处
《自动化学报》
EI
CSCD
北大核心
2001年第4期521-530,共10页
Acta Automatica Sinica
基金
National Science Foundation of P.R.China (60084002) and 863 High-Science Foundation for young researchers.
关键词
机械手
模糊聚类
离散滑动模控制
跟随控制
Algorithms
Fuzzy control
Manipulators
Mathematical models
Relay control systems
Sliding mode control
System stability
